vocabulary—into four "forms," our "nebula" being regarded as of the third. In 1785 Herschel informed Miss Burney that he had discovered fifteen hundred universes,—"fifteen hundred whole sidereal systems, some of which might well outvie our Milky Way in grandeur".
Such was the famous "disc-theory " and its corollary—the hypothesis of island universes. For many years this theory was expounded in text-books of astronomy and popular science manuals as if it had been the outcome of Herschel's matured views on the stellar universe. The late R. A. Proctor, who was one of the first close students of Herschel's papers, truly remarked that "It seems to have been supposed that his papers could be treated as we might treat such a work as Sir J. Herschel's 'Outlines of Astronomy'; that extracts might be made from any part of any paper without reference to the position which the paper chanced to occupy in the entire series". The consequence of this method of expounding Herschel's views was that for many years astronomers were hardly aware of his gradual change of opinion.
The disc-theory and its corollary were, as already noted, based on the assumption of an equal scattering of stars in the Milky Way and involved the belief that all nebulæ were stellar clusters which would ultimately be resolved into stars. The first of these views was never held very confidently by Herschel. In his paper of 1785 he admitted that "in all probability there may not be two or three of them in the heavens, whose mutual distance shall be equal to that of any other two given stars, but it should be considered that when we take all the stars collectively there will be a mean distance which may be assumed as the general one". Even in the paper of 1785 Herschel remarked that it would not be difficult to point to two or three hundred gathering clusters in our system. Indeed, his classification of so-called "nebulæ" was based on his view that condensation
4
